Most studies of the effectiveness of chemotherapy as a treatment for schizophrenia group all schizophrenics together, and also make use of some general criterion of the effectiveness of treatment. The present investigation is designed to test the hypothesis that newly admitted schizophrenic patients can validly be subdivided on the basis of objective tests of thought disorder into three subgroups: "process schizophrenics," "psychotic anxiety reaction," and "overinclusive psychotics." The different clinical response of the three schizophrenic subgroups to four weeks of phenothiazine treatment will be compared. The effect of the treatment on the psychological tests of thought disorder will also be studied, by retesting the patients following treatment. In addition, psychophysiological measures will be taken to test the hypothesis that "process schizophrenics," as defined, have an unusually low level of autonomic responsiveness as contrasted to those with "psychotic anxiety reaction."